INDEX

Explanations

information related to historical events or political figures

oai_token-act-pair · gpt-3.5-turbo

elements related to significant historical events or figures

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 NASCAR

-0.71

 Taco

-0.68

 Simmons

-0.66

 moms

-0.65

NBC

-0.65

 Tennessee

-0.64

Todd

-0.63

 NYPD

-0.63

 LAPD

-0.63

 Creep

-0.63

POSITIVE LOGITS

 organis

1.00

 favourable

0.96

 envis

0.91

 enqu

0.89

 organise

0.88

 analyse

0.88

 emphas

0.87

 analys

0.84

 organising

0.83

oubted

0.83

Activations Density 2.855%

information related to historical events or political figures

elements related to significant historical events or figures

No Comments

No Known Activations